We were the first to identify TGFBR1*6A, a common variant of the TGFBR1 gene (which encodes the type I TGF-beta receptor). We have shown that TGFBR1*6A transmits TGF-beta growth inhibitory signals significantly less effectively than does TGFBR1. In a meta-analysis of the three case-control studies published on TGFBR1*6A and breast cancer that include 555 cases and 1033 controls, 21.1% of cases and 13.6% of controls were TGFBR1*6A carriers and TGFBR1*6A was significantly associated with breast cancer (OR 1.48, CI 1.11-1.96). A common variant within the human TGF-beta1 (TGFB1) gene itself is represented by the substitution of Leucine to Proline (T-C) at the 10th amino acid position (T29C). The T-> C substitution results in higher extracellular TGFB1 secretion and higher circulating levels of TGFB1 have been observed in humans who carry the C allele. Investigation of this polymorphism with regard to breast cancer risk has yielded conflicting results. We propose to assess the individual and combined association between these two well-characterized and functionally relevant TGF-beta signaling pathway variants and breast cancer risk using a family-based association study comparing genotypes in 1844 women with breast cancer to those of their 2547 unaffected sisters. We have the following Specific Aims: 1) To assess the association between carrier status of the TGFBR1*6A allele and breast cancer risk. Hypothesis: carrier status of this allele is associated with increased risk of breast cancer. 2) To assess the association between the other functionally relevant variant of the TGF-beta signaling pathway, TGFB1 T29C and breast cancer risk. Hypothesis: carrier status of the T allele is associated with increased risk of breast cancer. 3) To assess the combined effects of TGFBR1 and TGFB1 variants that affect TGF-beta signaling on breast cancer risk. Hypothesis: Individuals with the lowest predicted levels of TGF-beta signaling, i.e. carriers of both the TGFB1 *T allele and the TGFBR1 *6A allele have a increased risk of breast cancer compared to individuals with the lowest predicted levels of TGF-beta signaling, i.e. homozygotes for both the TGFB1*C allele and the TGFBR1*9A allele.